docs-from-code

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs the agent to clone or download a user-provided remote repository and to run/consume the repo's files and graphify outputs (e.g., "If the user pointed you at a remote repo, clone it first" in SKILL.md and reading GRAPH_REPORT.md/source files), which are untrusted, user-generated third‑party contents that the agent must read and use to drive its actions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly installs and runs the external extraction tool "graphify" at runtime (pip install graphifyy and references https://github.com/safishamsi/graphify), which fetches/executes remote code and produces GRAPH_REPORT.md/graph.json that the agent uses to control prompts and generated docs.